INDEX
    Explanations
    New Auto-Interp
    Negative Logits
     descul
    -0.08
     renew
    -0.07
     despesas
    -0.07
    iaeth
    -0.07
    ované
    -0.07
     sixteen
    -0.07
    cem
    -0.07
    ues
    -0.07
    automatic
    -0.07
     separate
    -0.07
    POSITIVE LOGITS
     incons
    0.10
     mismatch
    0.09
     mism
    0.09
     inac
    0.09
     unmet
    0.09
     inconsist
    0.09
     inaccessible
    0.08
    Mismatch
    0.08
     κάποιο
    0.08
     irgendwo
    0.08
    Act Density 0.026%

    No Known Activations